LEDs luminous limits reached?

[metalhed]

Hardly.

New research at the National Institute of Standards and Technology (NIST) has given another possible method for increasing the luminous flux of an LED.

Novel Nano-Etched Cavity Makes LEDs 7 Times Brighter

Now it will take some time for this type of manufacturing trick to make it to production lines, but nevertheless, the future is bright. :hahaha:

 

[jtr1962]

Recently upgrading my bike light made me think about this thread. For about the last two years the light was using an array of 20 ChiWing 8,000 mcd LEDs running at 24 mA. Light output was about 50 lumens, and the array consumed about 1.5 watts, giving roughly 33 lm/W. The upgrade was done with an array of 28 Jeled 50Ks, which is the most that can physically fit in the space. The LEDs are run at 19 mA in two series strings of 14. Total power to the array is about 1.8 watts, output is around 150 lumens, and efficiency is about 83 lm/W.

While this improvement in light output is dramatic, I was thinking that this is probably the last time I will see a factor of three increase from a 5mm LED upgrade. For heat reasons the array power is limited to about 2 watts, and the number of LEDs is constrained to 28 for space reasons. Given that, another factor of three increase would require 28 LEDs outputting 450 lumens at 2 watts, or 225 lm/W. While this doesn't violate any of the laws of physics, I somehow doubt I'll see LEDs this efficient any time soon. 150 lm/W is about the best I can hope for, so at best any upgrade in the foreseeable future will perhaps only double output. Granted, with more efficient LEDs I can put more power to the array while generating the same waste heat, but even this doesn't help me much. I figured that the Jeleds are roughly 25% efficient, so with 1.8 watts to the array that's 1.35 watts of waste heat. Hypothetical 150 lm/W LEDs would be around 45% efficient, so I can run the array at about 2.45 watts instead. That still only gives me about 2.5 times the input of what I have now. And I doubt 150 lm/W LEDs will be available in 2 years in the same way that the Jeleds came out only two years after the ChiWings. It looks like it'll be a long wait even to double the output.

Sure, LEDs will continue to get brighter, but the only dramatic improvement on the horizon in the near future will be when we see power LEDs go from ~35 to 100+ lm/W. As for 5mm LEDs, in the near future we'll probably only go from 80 to 100 lm/W, an incremental and to most people's eyes barely noticeable increase.

 

[NewBie]

MIT's photonic lattice is making it's way to the market.

The part is known as the phlatlight, made by Luminous.
http://www.luminus.com/products/chipsets.php

The part is designed for 10-16 Amps, the emission area is small, and is currently designed for projection applications.


The next round of enhanced LEDs will be hitting the markets this fall and early winter, and the gains are very real. Meanwhile, LumiLEDs is still trying to get all the flavors of the K2 to the market, and it was delayed to the market, since it's announcement back in February 2005 (about a year and a half delay).

I hope there is something else comming down the pipeline from LumiLEDs that will be out this fall/winter, but I have heard absolutely nothing about any thing else they have which is inbound.

The next crop of parts is expected to make 2x the light with the same amount of power.

An example is a part made by Seoul Semiconductor, which they have been claiming will be in the 98 lm/W range. This is over double the output of other 1-5W high power LEDs.